In order to succeed in school, not only do children need good linguistic skills, but they also need to know how to use those skills to link ideas to one another-to make inferences, draw comparisons and analogies, construct hierarchies and partonomies, enlist schemas and definitions-in other words, they need to know how to use their language for higher order thinking. Higher order thinking has been identified as core to children's ability to become adaptive, innovative, and academically successful thinkers. Project 1 asks whether children vary in how they use talk for higher order thinking prior to school and, if so, whether this variability (1) predicts their higher order thinking later in development, and (2) is predicted by their parents' use of talk to illustrate and elicit higher order thinking earlier in development. The project builds on previously collected longitudinal observations of 60 children, ages 14 months to 9 years, whose families were chosen to represent the demographic range of Chicago. The parents in these homes have been shown to provide their children with widely varying amounts and types of linguistic input. The children will be followed until age 14 and, as they enter early adolescence, data will be collected on how they use language for higher order thinking in problem solving tasks and on standardized tests. Project I has three specific aims. (1) Study 1 describes changes in how children use connected discourse for higher order thinking from the eariiest stages of language learning (14 months) through early adolescence (14 years). (2) Study 2 explores how often parents use their talk to display and to elicit higher order thinking from their children in the early years (14-58 months) and later in development (10 years) when higher order thinking becomes particularly important for success in school. (3) Study 3 develops a model of cumulative parent input to higher order thinking that can be used to predict children's subsequent skills in higher order thinking. In addition. Project I will provide normative data for children with brain injury in Project II. The richness of the longitudinal data on which this project is based provides a detailed picture of the early talk children hear at home that could foster higher order thinking. RELEVANCE (See instructions): The goal of this project is to identify early parent behaviors that foster child higher order thinking. If these behaviors can be isolated, parents and preschool teachers can then be encouraged to use them when interacting with their children. The behaviors can also provide a basis for designing materials and interventions to promote higher order thinking, a skill that becomes more and more important as children progress through school.